In an industry where efficiency and innovation are paramount, Boris Cherny, the mastermind behind Claude Code, has unveiled a workflow that is turning heads in the tech community. His approach to software development is not only reshaping how code is written but also challenging traditional methodologies that have dominated for decades.
The Multi-Agent Paradigm
Cherny’s workflow is a masterclass in leveraging artificial intelligence to amplify productivity. By deploying multiple AI agents simultaneously, Cherny can manage what feels like a battalion of developers with the effort of just one person. His setup involves running five Claude instances in his terminal and another five to ten in the cloud, creating a dynamic environment that operates in real-time much like a strategy game.
This approach is not about replacing human developers but augmenting their capabilities. Each AI agent is tasked with specific functions such as testing, documentation drafting, and code refactoring, allowing for a seamless and efficient software development process. The result is a workflow that not only increases output but also enhances the quality of the code produced.
Choosing Intelligence Over Speed
One of the most intriguing aspects of Cherny’s workflow is his preference for using Anthropic’s Opus 4.5 model—a decision that flies in the face of the industry’s obsession with speed. This model, although slower, requires less intervention and steering, which ultimately reduces the correction time post-development. Cherny’s philosophy is simple: investing in smarter models upfront can save significant time and resources down the line.
Creating a Living Knowledge Base
A significant challenge in AI-assisted development is the issue of AI amnesia—where AI systems forget their previous mistakes. Cherny addresses this through a shared document, CLAUDE.md, which acts as a repository for all learned lessons and best practices. This living document ensures that every AI-generated error becomes a learning opportunity, fostering continuous improvement and adaptation across the development team.
Implications for the Future
Cherny’s approach not only sets a new standard for software development but also offers a glimpse into the future of AI integration in tech. By demonstrating that AI agents can work in harmony with human developers, Cherny is paving the way for more innovative and efficient workflows that could redefine productivity standards across industries.
Conclusion
The introduction of Cherny’s workflow is more than just a technical revelation; it’s a cultural shift in how we perceive and execute software development. As companies strive to do more with less, the principles established by Cherny could become foundational in building the next generation of development tools. The implications of his work could well transcend the confines of coding, influencing broader technological and organizational strategies.
